Posteroanterior movements in tender and less tender locations of the cervical spine

Abstract

In order to determine how posteroanterior movements (PAs) are related to tenderness and thus possibly symptom production, we measured PA movements to a force of 25 N on each side of the cervical spines of asymptomatic subjects. From 10 subjects (six females and four males; mean age 37.2, range 21–50), 10 locations with a difference in tenderness to pressure between sides were used for analysis. The force–displacement and stiffness–force curves for tender and control sides were compared in four ways: simultaneous confidence bands (SCBs) for each side; width of SCBs for each side; SCBs of the difference between pairs of the tender and control curves; and simultaneous prediction bands (SPBs) from the tender side were compared to individual curves of the controls. The tender side demonstrated greater variation of both displacement and stiffness. The tender sides demonstrated greater within-subject stiffness for all force levels above 12 N. All individual stiffness–force curves of the tender sides were significantly different from the control side. Expected differences in single measures of either displacement or stiffness were not detected. The results suggest that the pattern of stiffness is a more effective method of characterizing PA mobility than single measures used in previous studies.

Introduction

Musculoskeletal symptoms such as neck or back pain are amongst the most common reasons patients seek medical attention (Bogduk et al., 2003). Dysfunction of movement between individual intervertebral motion segments is considered to be a potential source of spinal musculoskeletal symptoms (Banks, 1998) and passive movement tests such as spinal posteroanterior movements (PAs) are intended to localize and assess the dysfunctional intervertebral movement (Maitland et al., 2005).

Although many authors advocate motion palpation as an important component of physical examination (Bullock-Saxton et al., 2002), the usefulness of passive movement tests such as spinal PAs has been brought into question by inconsistent repeatability (Smedmark et al., 2000; Pool et al., 2004). In spite of a lack of repeatability, manual assessment of passive movement has been shown to be useful clinically. For example, symptomatic locations (Jull et al., 1988) and the location of congenital fusion have been reliably detected by manual motion palpation (Humphreys et al., 2004). In a clinical study, the lumbar spines of patients were classified by findings on manual palpation as hypomobile or hypermobile. Patients who received corresponding treatment (manipulation to increase segmental mobility for the hypomobile group and stabilization exercises to counteract excessive mobility for the hypermobile group) had better treatment outcomes than those receiving randomly allocated treatment (Fritz et al., 2005).

Spinal PAs were previously thought to produce isolated movement between a target pair of vertebrae and the response felt by the therapist was considered to be a direct indicator of the local intervertebral movement (Grieve, 1981). It is now clear from both in vivo (Lee and Evans, 1997; Caling and Lee, 2001; Kulig et al., 2004; Lee et al., 2005) and in vitro (Gal et al., 1997; Sran et al., 2005) studies that in addition to moving the target intervertebral segment, spinal PAs also move other structures including a number of intervertebral levels as well as extra-spinal structures.

In order to clarify the usefulness of spinal PAs, a number of investigators developed instrumented methods to objectively assess PA movements. Interpretation of data from studies discussed in a recent review (Shirley, 2004) relied on single scalar values of displacement or stiffness extracted from the force–displacement (FD) curves of spinal PAs to characterize the stiffness of the entire movement. Using these single values to assess stiffness, instrumented measures of spinal PAs have been successful in detecting differences occurring with segmental dysfunction. For example, differences have been demonstrated with reduction in symptoms in patients with low back pain (Latimer et al., 1996b), artificially induced disc degeneration in a porcine model (Kawchuk et al., 2001), and local intervertebral stiffness in vitro in human thoracic spines (Sran et al., 2005). Using the same criteria for assessing stiffness, differences have been found also in relation to a wide variety of factors whose influence is extraneous to local intervertebral mobility (Kawchuk and Fauvel, 2001) including subject position (Edmondston et al., 1998; Chansirinukor et al., 2001), stage of respiration (Shirley et al., 2003), size of the indentor (Squires et al., 2001), and muscle contraction (Hodges et al., 2003; Colloca and Keller, 2004). The methodologies used in these studies are able to detect altered stiffness of spinal PAs resulting from a variety of structures, but are unable to differentiate between alterations resulting from the targeted intervertebral segment and those resulting from extraneous factors.

We therefore set out to identify particular patterns of spinal PA stiffness associated specifically with intervertebral dysfunction (as indicated by local tenderness to pressure) using a protocol similar to that recommended for manual assessment of unilateral PAs (Maitland et al., 2005). That is, we compared PA movements at tender and less tender locations that would otherwise be expected to be as similar as possible; i.e. side-to-side at the same spinal level. Rather than relying on single values of stiffness or displacement, we compared the patterns of displacement and stiffness throughout the PA movement using a bootstrapping method of calculating simultaneous confidence bands (SCBs) and simultaneous prediction bands (SPBs) to detect more specific differences. We hypothesized that, in addition to reduced displacement and an increase in single values of PA stiffness, more specific differences in the patterns of stiffness throughout the PA movement would correspond to differences in local tenderness to pressure.

By understanding the specific characteristics of PA stiffness related to tenderness (and presumably intervertebral dysfunction), we hope to enable more accurate interpretation of manual and instrumented assessment of PA movements.